• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining;International Symposium on Foundations of Open Source Intelligence and Security Informatics;International Symposium on Foundations and Applications of Big Data Analytics;International Symposium on Network Enabled Health Informatics, Biomedicine and Bioinformatics >Followee Management: Helping Users Follow the Right Users on Online Social Media
【24h】

Followee Management: Helping Users Follow the Right Users on Online Social Media

机译：追随者管理：帮助用户在在线社交媒体上追随正确的用户

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

User timelines in Online Social Media (OSM) remains filled with a significant amount of information received from followees. Given that content posted by followee is not under user's control, this information may not always be relevant. If there is large presence of not so relevant content, then a user may end up overlooking relevant content, which is undesirable. To address this issue, in the first part of our work, we propose suitable metrics to characterize the user-followee relationship. We find that most of the users choose their followees primarily due to the content that they post (content-conscious behavior, measured by content similarity scores). For a small number of followees, a high degree of social engagement (likes and shares) irrespective of the content posted by them is observed (user-conscious behavior, measured by user affinity scores). We evaluate our proposed approach on 26,516 followees across 100 random users on Twitter who have cumulatively posted 234,403 tweets. We find that on average for 60 % of their followees, users exhibit very low degree of content similarity and social engagement. These findings motivate the second part of our work, where we develop a Followee Management Nudge (FMN) through a browser extension (plugin) that helps users remain more informed about their relationship with each of their followees. In particular, the FMN nudges a user with a list of followees with whom they have least (or never) engaged in the past and also exhibit very low similarity in terms of content, thereby helping a user to make an informed decision (say by unfollowing some of these followees). Results from a preliminary controlled lab study show that 62.5 % of participants find the nudge to be quite useful.
机译：在线社交媒体（OSM）中的用户时间表仍然充满着从关注者那里收到的大量信息。鉴于关注者发布的内容不受用户控制，因此此信息可能并不总是相关的。如果存在大量不那么相关的内容，则用户可能最终会忽略相关内容，这是不希望的。为了解决这个问题，在我们的工作的第一部分中，我们提出了合适的度量标准来表征用户与追随者的关系。我们发现，大多数用户选择他们的关注者的主要原因是他们发布的内容（通过内容相似性评分衡量的内容意识行为）。对于少数追随者，无论他们发布的内容如何，​​都可以观察到高度的社交参与度（喜欢和分享）（用户自觉行为，通过用户亲和力得分衡量）。我们对Twitter上100个随机用户中的26,516个关注者评估了我们提出的方法，这些用户累计发布了234,403条推文。我们发现，平均而言，在60％的关注者中，用户的内容相似度和社交参与度非常低。这些发现激发了我们工作的第二部分，在该工作中，我们通过浏览器扩展（插件）开发了“追随者管理微调”（FMN），该扩展程序可以帮助用户更了解自己与每个追随者的关系。尤其是，FMN会向用户提供其与过去（或从未）有过最少（或从未与之互动）的关注者列表，并且在内容方面也表现出非常低的相似性，从而帮助用户做出明智的决定（例如，通过取消关注其中一些追随者）。初步的受控实验室研究结果表明，有62.5％的参与者认为微调非常有用。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号